Why don't tumor-reactive T cells eliminate tumors? Although priming in the draining lymph node is often suboptimal, more frequently, the immune response is stymied at the tumor itself. This may be as a result of a protection of the tumor from immune surveillance (tolerance) or inadequate repriming and feedback to the local draining lymph node (an interrupted feedback loop). Based on a panel of preliminary data, we hypothesize that a subpopulation of APCs in the tumor microenvironment protect the tumor from CTLs by direct interaction with the T cells, by their failure to mature in response to the interaction, and by cooperation with regulatory T cells. We believe the microenvironment builds an arsenal of these cells and collects the immune system at 'inactivating foci'that prevent stable and effective surveillance of the tumor by primed CTLs. By simultaneously preventing the I cells from seeing the tumor and suboptimally activating them there, these cells may effectively neutralize the response. We propose to study this in a new model system of human breast cancer, the PyMTCherry-OVA mouse that we have produced and characterized expressly to address this poorly accessible problem. The model allows direct visualization of the tumor AND the tumor-sampling APCs via intrinsic fluorescence. This is significant as we are able to study what is happening to the T cell as it contacts these distinct cell types, specifically via 2-photon microscopy. It is also significant because we are also able to phenotype this very important APC population using multicolor flow cytometry, microarray, and in vitro assays and provide key data about the co-development of APC and I cell population that results in tumor acceptance. It should be noted that the studies proposed are amongst the first of their kind that will be achievable in a non-ectopic tumor model that very closely resemble human breast cancer in many important ways. The spontaneous eteliology of our model is also important, as there are many reasons that ectopic tumors are unlikely to faithfully replicate important aspects of immune-evasion, including the microenvironment, by a primary tumor. Thus, studies in this system are much more likely to recapitulate the features of human disease and the 'normal'establishment of immune evasion. The results of these experiments are going to reveal the dynamics of I cells over time in the developing hyperplasia through to established carcinomas. They will also definitely reveal the interaction of regulatory I cells with cells of the local I cell response. Finally, and not insignificantly, they will shed considerable light upon a previously inaccessible cell population which may protect the tumor from the immune response and thus permit tumor growth and metastasis.